Project Summary: Gene activity is modulated in different cell types by the way genomic DNA is packaged into chromatin ? a process termed ?epigenetics?. Epigenetic controls are disrupted in nearly all forms of cancer, as well as in many other human diseases. This may occur through mutation of chromatin regulators, environmental exposures that alter chromatin structure, or inappropriate developmental cues. There is enormous enthusiasm for the potential of ?epigenetic therapies? to correct epigenetic defects in clinical settings. However, we currently lack coherent models or mechanistic understanding of how epigenetic defects promote tumors, or how they might be modulated in clinical intervention. The proposed project will pursue a novel, unifying model for how epigenetic lesions drive tumor initiation and evolution. We hypothesize that a key function of most epigenetic lesions is to induce plasticity, which allows pre-malignant or malignant cells to stochastically sample alternate gene regulatory programs. Cells that adopt programs that confer fitness (proliferation, tolerance, etc) are selected, and their epigenetic state maintained through cell division, giving rise to a new lineage and, ultimately, to malignant progression. Newly established technologies for profiling, monitoring and modulating epigenetic landscapes, including at single cell level, provide a unique opportunity to test this hypothesis and characterize the underlying mechanisms. We will focus initially on exemplar lesions that drive brain tumor initiation and evolution. The first exemplar is isocitrate dehydrogenase (IDH) gene mutations and associated DNA hyper-methylation, which we hypothesize cause stochastic disruption of chromatin boundaries and insulators, thereby allowing aberrant induction of oncogenes. The second exemplar is stress-induced histone demethylation, which we posit allows cancer stem cells to access primitive developmental programs and evolve drug tolerance. We will deeply characterize these exemplars by leveraging clinical specimens and experimental models, and by further innovating new technologies. We will then explore the extent to which plasticity pertains to other oncogenic lesions and to other diseases with epigenetic etiologies. In summary, the proposed study will deeply investigate mechanisms of epigenetic deregulation and plasticity in tumorigenesis. The research has potential to radically alter current views of epigenetic regulation in human disease, and thus has important biomedical implications.